Title: How the brain dances: EEG analysis of the creation, teaching, learning and performing phases of dance
Domain Experts: Anthony Roberts (Dance), Jeremy Loebach (Psychology)
MSCS Mentor: Matt Richey
CIR Fellows: Johanna Glaaser, Noah Hillman, and Katie Voegele
Description: The primary focus of this project is to create an analysis platform for brain activity during the creation, teaching, learning and performance of dance sequences. This research was conducted as a 2017 CURI collaboration between Anthony Roberts (Dance), Jeremy Loebach (Psychology and Neuroscience) and students Fiona Steen and Randall Rude. Using an open source wireless EEG system (Open BCI), neural responses from 8 scalp locations were recorded from three participants of varying expertise (novice, intermediate, and professional) as they created, taught, learned and performed novel dance pieces. The project sought to better understand what measurable brain activity occurs when a novel movement sequence is created, communicated, learned and executed, and how experience contributes to such activity. Our initial analysis showed marked differences between phases across all three participants, but as often happens with summer projects, we simply ran out of time to dig into the data more thoroughly.
Students will be responsible for the analysis of EEG data from 8 scalp locations across 3 dancers during 4 phases of a dance activity, across 10 individual sessions. Data currently exist in csv format, and may be analyzed in Matlab or other suitable programs. We are particularly interested in tools for data normalization to allow the comparison across different electrode conditions across trials, across participants, across sessions and across days. We are also interested in looking for commonalities across electrode locations and phases as the experiment progressed. While the main goal for this project is to develop analysis tools that will be applied in the current data set to answer the earlier mentioned research question, we anticipate that these tools will be used in future projects by faculty and students from the Fine Arts and the Sciences.
Title: Fossil Coral and Ocean Carbon Cycling
Domain Expert: Anne Gothmann (Physics and Environmental Studies)
MSCS Mentor: Paul Roback
CIR Fellows: Spencer Eanes, Penelope Lancrete, and Emily Patterson
Description: Records of past climate and carbon cycling on Earth can help elucidate how Earth’s climate will evolve in the future. Recently, the uranium to calcium ratio (U/Ca) of fossil coral skeletons has been identified as a potential tool for reconstructing past ocean pH, which in turn reflects ocean carbon cycling. Laboratory experiments generally support the relationship between coral U/Ca and seawater pH, but other studies suggest that U/Ca in corals may depend on seawater temperature and coral species as well.
Building on work done by CIR students in 2018-2019, we will test the hypothesis that fossil coral U/Ca can be used to reconstruct ocean pH. We will also explore whether there are other environmental parameters (for example, temperature or salinity) that influence U/Ca ratios in coral skeletons. Our approach will be to first use data manipulation and management skills to synthesize and compile existing data on (1) the U/Ca ratio measured in coral skeletons and on (2) seawater conditions (including seawater temperature and pH) in which coral skeletons were grown. Then, through statistical analysis, we will determine the environmental factors that are primarily responsible for observed variations in the U/Ca ratio of coral skeletons.
Title: Pond Thermal Profiles and Mixing Patterns
Domain Expert: Meredith Hogerson (Biology and Environmental Studies)
MSCS Mentor: Joe Roith
CIR Fellows: Alana Barnhart, Kshitij Gurung, and Andy Ness
Description: Ponds vastly outnumber larger lakes worldwide, yet ponds are understudied and we know little about patterns driving pond chemistry, biology, and physics. One key feature of lakes and ponds is their temperature structure across the water column, which we call thermal profiles. Lakes tend to have stratified thermal profiles, where the bottom water is cold and the water on top is warm. Lakes typically mix twice a year: once in the spring and once in the fall, during which temperatures are the same at the top and bottom. In ponds, however, mixing is much more common. This mixing affects many chemical and biological phenomena in ponds, including nutrient recycling, primary production, greenhouse gas emissions, and oxygen dynamics. Yet, we do not know how often ponds mix versus stratify, nor do we understand the drivers of pond mixing.
In this project, we will examine thermal profiles and mixing patterns in ponds from around the world. This is part of a global collaboration to understand pond dynamics. The data include high-frequency temperature data from throughout the water column of 50-75 ponds located around the world. We will identify the strength of stratification, duration of stratification, and drivers of mixing, including pond characteristics (e.g., size, depth, water clarity) and weather data.
Title: Ten Years Later: Revisiting Chemistry Placement Exam Analysis
Domain Experts: Jeff Schwinefus and Mary Walczak (Chemistry)
MSCS Mentor: Katie Ziegler-Graham
CIR Fellows: Megan Hussey, Maggie Upjohn, and Owen Wright
Description: The St. Olaf Chemistry Department has used a placement exam for many years to guide incoming students into an introductory chemistry course best suited to their background and success in first-year chemistry. Ten years ago we undertook a revision of the placement exam through a CIR Project. Now, we seek to re-examine and update our placement process. Our current process relies on student scores from an online chemistry test along with admissions data such as High School GPA and standardized test scores. As we look to the next decade of chemistry placement we want to ensure that we are using the best process to place students into Chem 121, 122, 125 and CH/BI 125. As more students matriculate without high school GPAs (e.g., international students) and many colleges eliminate the standardized test requirement for admission, we wonder whether these data are critical to placing students into our chemistry courses.
Title: Doctor Behavior around End-of-Life Counseling
Domain Expert: Ashley Hodgson (Economics)
MSCS Mentor: Sharon Lane-Getaz
CIR Fellows: Victoria Knutson, Tyler Radtke, and Rayan Sadeldin
Description: The U.S. Medical system is sometimes accused of plugging forward in aggressive treatment of terminally ill patients without first consulting the patients about their own values and priorities. End-of-life conversations can be reimbursed by Medicare in a number of ways, including advanced care planning with the doctor, palliative care consultation, and hospice consultation. But little is known about the characteristics of doctors who are most likely to initiate these conversations, and how early in a patient’s illness the doctor initiates. Do doctors who financially benefit from life-extending treatments have these conversations later or not at all? Does the race or gender of the doctor (possibly in relation to the patient) matter for the doctor’s likelihood of initiating?
Using a 100% “sample” of Medicare patients and their doctors for 2016, we hope to identify the characteristics of doctors who are most likely to have conversations with patients well ahead of their death. Among patients in our data set who died and had an end-of-life counselling session with a doctor prior to their death, how close did that conversation happen to their date of death? We can classify doctors according to the “typical” lead time before death that this conversation takes place. We will then compare doctors who have conversations well ahead of the patient death (>6 months) to doctors who generally initiate this conversation once the patient’s death is more imminent. Students on this project should be willing to research and learn machine learning algorithms for identifying patient medical conditions (as indicated through ICD-10-CM codes) most associated with impending death.